Method for determining period of oestrum in animals for insemination

Abstract

FIELD: agriculture.SUBSTANCE: method for determining the period of oestrum in animals for use in insemination. A pen containing a group of sows is enclosed, and the sows in this group are potentially in oestrum. An image of the vulva of a sow from a specified group of sows is obtained periodically. The obtained image of the vulva is analyzed using a control system to conclude from the image whether the sow is in oestrum. A database is formed from a set of images showing the vulva of a set of sows known to be in oestrum and images showing the vulva of a set of sows known to be not in oestrum. In this case, the said conclusion is based only on a program that uses the obtained image in comparison with the said database, and an output signal is generated regarding the location of the sow in the first period of oestrum in response to the said conclusion.EFFECT: automation of determining the period of oestrum in animals is increasing.20 cl, 7 dwg

Description

[0001] The present invention relates to a method for determining the period of estrus in animals for use in insemination.

[0002] Prerequisites for the creation of the invention

[0003] In order to effectively manage livestock farming, much attention has been paid to finding effective ways to determine when the animal in question is in heat so that insemination in such a situation can be carried out with a higher probability of success.

[0004] It is known that changes in mobility and overall behavior, including changes in the ratio between periods of standing and periods of lying down, can be used to determine the period of estrus. Furthermore, body temperature, particularly that of the vaginal area, can be used for this determination.

[0005] However, further improvement of methods for determining the period of estrus is required.

[0006] The present invention is particularly applicable to sows, where workers typically have less contact with the animals, requiring a more automated method to determine the heat cycle. However, this method can also be applied to other animals.

[0007] Disclosure of the essence of the invention

[0008] The object of the present invention is a method for breeding livestock, comprising:

[0009] restriction of the area of ​​a pen in which a group of sows is located, with the sows potentially in heat;

[0010] preparation of a boar, which is located in an isolated or fenced area (pen) or cage;

[0011] The said pen and cage are constructed in such a way that the sows in the pen can come into close contact with the boar when the boar or sow is moved;

[0012] at least one collector device for collecting images and / or sounds from each sow when it is in said close contact;

[0013] and analyzing images and / or sounds using the control system to determine whether the sow is in heat for insemination.

[0014] The method proposed herein may thus use either images captured by the camera or sounds emitted by the sow, or may combine the analysis of both sets of data collected when the sow is in close contact with the boar.

[0015] In a preferred embodiment of the method, images are obtained by photographing the sow from the front to show the body and head, as these indicate changes in mobility and behavior that can be used to determine the onset of estrus. Specifically, frontal images of the sow can be taken while the sow is in front of and very close to the boar, a position in which the changes are expected to be most noticeable.

[0016] In a preferred embodiment of the method, images are analyzed to detect sow movements characteristic of estrus. Such characteristic movements are determined by preliminary analysis of the movements of sows known to be in estrus and comparison with the currently acquired image or images. To obtain more accurate results, a large number of comparisons can be performed using statistical analysis to eliminate errors.

[0017] In particular, images can be analyzed to detect characteristic features known as indicators of estrus, including:

[0018] speed of movements characteristic of the period of sexual hunting.

[0019] Body movements characteristic of the period of sexual hunting.

[0020] Head movements characteristic of the period of sexual hunting.

[0021] Ear movements characteristic of the period of sexual heat.

[0022] angular velocity of ear movement, characteristic of the period of sexual hunting.

[0023] movements of the dorsal region characteristic of the period of sexual hunting.

[0024] changes in body posture characteristic of the period of sexual hunting.

[0025] In particular, sounds can be analyzed to detect characteristic signs known to indicate the onset of heat, including:

[0026] specific types of sounds emitted by a sow and characteristic of the period of estrus, such as squeals, grunts.

[0027] frequency spectrum of sounds characteristic of the period of sexual hunting.

[0028] a sequence of sounds characteristic of the period of sexual hunting.

[0029] Preferably, the collecting device comprises a camera mounted on the apparatus into which the boar is placed and directed toward the sow located near the apparatus.

[0030] In one embodiment, the apparatus in which the boar is placed contains a movable structure, and the boar is transported to sows located in individual pens. Typically, the apparatus is a trolley, which holds the boar in a cage and moves along passages to sows located in pens located along these passages. A boar trolley is the best way to restrain and control a large boar as it moves from sow to sow. However, other means can be used for boar transport, such as a simple tether or a robotic guide device, to which the boar is tethered as it moves from sow to sow.

[0031] In an alternative configuration, the boar isolation or restraint apparatus is fixed in place, such as a fixed cage in an open pen, where sows are free to move from place to place and also to approach the boar, which is separately confined in its own cage within that pen.

[0032] In both cases, the camera is preferably positioned on a trolley or on a cage and faces the sow to be imaged, so that a series of images of the sow can be obtained as it approaches or has already approached, in order to use these images in the analysis of the sow's movements as defined above.

[0033] A directional microphone is also preferably installed on the boar cage to mainly pick up sounds emitted by the sow that reach the boar cage.

[0034] While the above statements contain the term "boar," it is clear that it could be an actual live animal. However, such a boar could be an artificial construct that provides the sow with one or more stimuli that mimic a live animal. In extreme cases, such a "boar" could simply involve the use of a pheromone spray to adequately simulate a real live boar. It has been found that a live animal is more effective in stimulating the sow to detect estrus and accept insemination. However, improvements in stimuli simulating the boar could eliminate the need for transporting or cage-keeping a real, live, very large animal.

[0035] In one important embodiment, which can be used with any of the above theses or any of the properties defined below, the control system can distinguish between cases of heat determination requiring simple analysis and cases requiring complex analysis. In the case of simple analysis, an output signal is generated indicating the result of heat determination in a sow, while in the case of complex analysis, an output signal is generated requiring additional intervention before heat determination is made. In other words, the system can generate and indicate the level of confidence in the result, which allows the system to proceed to heat determination or to indicate that the confidence is low and that additional data is required. For example, additional data could come from the methods described below or from other worker intervention.As another example, a worker may be used to apply pressure to the dorsal region of a sow while collecting sounds and / or images.

[0036] The result of performing such a determination is that insemination is performed after receiving a signal indicating that a final determination has been made, so that the success rate of the insemination procedure is increased.

[0037] As another method that can be used herein as a separate system or part of the above analysis, a method for breeding livestock is provided, comprising:

[0038] creating a confined pen containing a group of sows, where these sows are potentially in heat;

[0039] Periodic collection of rear profile images of each sow;

[0040] Analyze each image to determine the area in that image that shows the sow's vulva;

[0041] and analyzing, using the control system, images of the said area to determine, based on features related to the shape, size and / or color of the vulva, whether the sow in question is in heat;

[0042] Generating an output signal in response to determining that the sow in question is in heat;

[0043] Here insemination can be performed after receiving a signal indicating that the estrus period has been determined.

[0044] Determining whether a sow is in heat can be used to determine the timing of insemination or simply to determine whether the animal is in heat. In some cases, there are sows for which the farmer wants to skip the heat cycle to optimize reproductive results;

[0045] According to a preferred method, the analysis is performed by a control system using a program that pre-analyzes a large number of images showing the vulval region of a plurality of sows known to be in estrus and images showing the vulval region of sows known to be not in estrus to generate data based on these previous images in order to create a program that generates a signal for determining the estrus period using this data. Thus, this program running in the control system is a neural network pre-trained with characteristic features for determining the presence of estrus by analyzing a large number of images of the vulval region of animals known to be in estrus and animals known to be not in estrus.This pre-trained program can be stored in a control module, which can be located directly on the user's site or in the cloud, for application to new images that need to be analyzed.

[0046] While the above machine learning is performed by the system itself, the present inventors, through advanced image analysis, have discovered that heat determination can be performed based on, for example, certain visual features that can be extracted from images. These features may include one or more of the following:

[0047] -a- the texture of the vulva or the wrinkles on it; this is related to the degree of swelling of the vulva, since the absence of wrinkles is a consequence of the increase in volume caused by the overflow of blood, which stretches the skin, reducing the depth and appearance of wrinkles.

[0048] -b- the height or width of the vulva; this is also related to the degree of swelling of the area, as the height and width of this area increases as a result of swelling.

[0049] -c- vulva color; the presence of engorged blood leads to a deepening of the red color that can be observed.

[0050] -d- Outwardly directed protrusions of the vulva. Again, the overflow of the area due to increased blood flow will act to stretch the vulva away from the body, generating an enhanced three-dimensional (3D) effect, which can be observed even in a two-dimensional (2D) image taken or selected at an appropriate angle.

[0051] In some cases, particularly in older animals that have produced multiple litters, changes in the vulval area may be less noticeable, so analyzing a single image may not allow for a definitive determination. In such cases, comparison with previous images of the same animal can be used to observe changes over time. At least one image of the same area is compared with earlier images of the same animal to utilize the time difference for determination. In other words, for a large number of sows, changes are sufficiently pronounced that the presence of estrus can be detected from a single image or from a series of images taken at the same time, so that they are representative of the current condition during the test.It can be understood that image capture and identification are performed repeatedly over a period of time, for example, daily, to determine the onset of estrus. This creates a series of images for comparison over a period of time. In other cases, determining estrus requires comparison with older images. However, this obviously requires that the images be associated with the animal's identity, which is easily accomplished if the animal is isolated, but requires additional steps if the animal is freely moving among other animals.

[0052] The first stage of the analysis preferably involves acquiring and analyzing multiple rear profile images to select one or more suitable images that best depict the desired area. When the camera is positioned and designed in such a way that its viewing angle and position relative to the animal's rear end can vary, many rear profile images may be unsuitable for depicting the vulva area. Therefore, the control system uses software to select only the appropriate images from the multiple images and use these suitable images to determine the estrus period. As part of this analysis, the system must select only the relevant vulva area from the entire rear profile image, as part of the algorithm requires first identifying this area and selecting it from the image to enable further analysis.In other words, after selecting the vulvar region, it is necessary to select from the subset of component images those that best depict this region and should be used to determine the estrus period. The system then evaluates each image to determine whether it indicates a positive or negative result. After this determination, a comparison of all positive and negative results will lead to a final status determination and will display a confidence level indicating how many results agree. This information can be made available for assessing whether insemination should be performed—in other words, whether the vulva shows signs of estrus—or whether further analysis with additional data is required.

[0053] After the first analysis to select from the row posterior profile images a subset of images that best or appropriately show the vulvar area, this plurality of images are selected and compared to obtain a determination result using statistical analysis.

[0054] In some embodiments, sows are housed in individual pens, and images are captured using a camera mounted on the pen. In this case, the camera may be mounted in a fixed position on the pen and synchronized and activated to capture rear profile images repeatedly at a frequency determined by the system, such as daily. The camera can then capture multiple rear profile images for the analysis described above. This takes into account that the animal may not be in the appropriate position or angle relative to the camera to capture the best rear profile images. In other cases, images are captured by a moving camera (mounted on a track, controlled by a robot, etc.), so not all such images may be suitable.

[0055] Posterior profile images are preferably taken with the animal standing, as this ensures that the vulvar region is correctly represented posteriorly and places this region of the vulva in a specific position on the image, facilitating selection of the desired images. Furthermore, this standing position ensures that the regions of this region are represented consistently on the image, allowing for more efficient analysis.

[0056] It may therefore be desirable to ensure that the animal is standing while images are being taken, or even to induce the animal, through some form of physical stimulation, to move into a standing position for the image to be taken.

[0057] In some embodiments, sows are not confined to individual pens and can roam freely. In this case, images can be captured using a camera mounted in a suitable location, such as a feeder or waterer frequently visited by the sows. This could be a boar display station with a window so that the sows can see the boar to achieve sexual arousal. With this approach, each sow enters the camera's field of view at random times and can be photographed each time she enters the area. In one embodiment, it is necessary to apply an identifier to the animal being photographed so that it can be identified for insemination. Therefore, in this case, animals can be tagged with a readable tag, such as a radio-frequency identification (RFID) tag attached to the ear. This identifier can also be used for the historical data comparison described above.

[0058] Alternatively, when identification information is unavailable, a sow can be marked after readiness is determined, which serves as the target for subsequent insemination. Such marking systems, using paint spraying or stamps, are available and can be activated to mark the sow in question. This is particularly effective when the filming camera is positioned in a feeding pen with walls that restrict one animal at a time. This allows only a single animal to enter the pen, be confined by the walls in a suitable orientation for filming, photographed, and then determine its estrus period before returning to the herd. Alternatively, a path can be created between two areas, such as a resting area and a feeding area, where the animal is temporarily held until the necessary procedures are completed before moving on to the next area.

[0059] In some cases, the system may be implemented as a program that can be applied to an existing image acquisition system, such as a mobile phone. In this case, images can be captured manually by a worker holding the imaging device. The worker provides information about the animal in question to the data storage system and selects the images to be captured. From these images, the control system selects images of the specified area of ​​interest for use and, based on these selected images, determines the estrus period for transmission to the worker. In other cases, the control system may be part of a stand-alone system containing a camera, a processor for determining the estrus period, and an interface for communication with the worker. In both cases, the control system can be designed to provide data regarding the estrus period determination for multiple sows to the sow housing system.

[0060] As explained in detail above, the proposed method of using images can be applied in combination with providing a boar that is confined in a pen, wherein the area of ​​the pen where the sows are located and the system of the pen that confines the boar are arranged so that the sows in the pen can come into close contact with the boar as a result of the movement of the boar or sows.

[0061] Brief description of drawings

[0062] One embodiment of the present invention will now be described in conjunction with the accompanying drawings, in which:

[0063] Fig. 1 is a simplified plan view of a barnyard in a livestock farm, showing the configuration used in the first embodiment of the method according to the present invention.

[0064] Fig. 2 is a simplified plan view of a barnyard in a livestock farm, showing the configuration used in the second embodiment of the method according to the present invention.

[0065] Fig. 3 is a logical diagram showing the steps of the first embodiment of the method according to the present invention.

[0066] Fig. 4 is a simplified plan view of a barnyard in a livestock farm, showing the configuration used in the third embodiment of the method according to the present invention.

[0067] Fig. 5 is an illustration of one example of an image of an animal in a rear view used in a third embodiment of the method according to the present invention.

[0068] Fig. 6 is an illustration of one example of an image of a region of an animal taken from Fig. 5 and used in a third embodiment of the method according to the present invention.

[0069] Fig. 7 is a logical diagram showing the steps of a third embodiment of the method according to the present invention.

[0070] Implementation of the invention

[0071] Figure 1 shows the components used in the livestock breeding method. The pen area 10 contains several mating sections 11 arranged side by side for isolating sows. These sections are arranged in a row along an alley 14.

[0072] The sections are formed by section walls made of welded rods with a 12-piece feeder along the front wall, so that the animals remain isolated in these sections during the breeding season. A group of sows is placed in these sections, with these sows potentially in estrus.

[0073] The boar is placed on a movable carriage 15 on wheels 16 so that the carriage can move on these wheels along the passage 14 and approach each sow in turn, so that each sow can come into close contact with the boar as a result of the movement of the boar. A suitable carriage is shown in earlier U.S. patents: U.S. 6,196,975 (Labrecque), issued March 6, 2001, and U.S. 6,508,202 (Kleinsasser), issued January 21, 2003. Both patents belong to the present applicant, and the disclosures of these patents are incorporated herein by reference.

[0074] The trolley carries a camera 17 and a microphone 18, mounted in a fixed position on the trolley. The camera captures images and the microphone collects the sounds emitted by each sow when she is in close contact with the boar. The camera and microphone are positioned in a suitable location on the trolley frame so that they are aimed at the desired position when the trolley is aligned with the sow section.

[0075] The resulting images and sounds are transmitted by wire or radio to the control system 19, which analyzes the data as described below to determine whether the sow is in estrus for insemination.

[0076] As shown in the logic diagram presented in Fig. 3, the proposed method uses either camera images or sounds emitted by the sow, or can combine the analysis of both data sets, where images and sounds are collected when the sow is in close contact with the boar. The steps of the method are shown in detail in Fig. 3.

[0077] According to the preferred method, images are taken from the front of the sow, showing the body and head, as these parts are indicative of changes in mobility and behavior, which in turn help determine the state of estrus. Specifically, images of the front view of the sow can be taken from the front, with the sow in close proximity to the boar's anterior region, where changes are expected to be most noticeable.

[0078] As shown in Fig. 3, these images are analyzed to identify sow movements characteristic of estrus, which were identified in a previous analysis of sows known to be in estrus, and the currently obtained image or images are compared with the database. To obtain more accurate results, multiple comparisons can be made using statistical analysis to eliminate errors.

[0079] In particular, images can be analyzed to detect characteristic features known as indicators of estrus, including:

[0080] speed of movements characteristic of the period of sexual hunting,

[0081] body movements characteristic of the period of sexual hunting,

[0082] Head movements characteristic of the period of sexual hunting,

[0083] ear movements characteristic of the period of sexual hunting,

[0084] angular velocity of ear movement, characteristic of the period of sexual hunting,

[0085] movements of the dorsal region characteristic of the period of sexual hunting,

[0086] changes in body posture characteristic of the period of sexual hunting.

[0087] In particular, sounds can be analyzed to detect characteristic signs known to indicate the onset of heat, including:

[0088] specific types of sounds emitted by a sow and characteristic of the period of estrus, such as squeals, grunts,

[0089] spectral composition (frequencies) of sounds characteristic of the period of sexual hunting,

[0090] a sequence of sounds characteristic of the period of sexual hunting.

[0091] As shown in Fig. 2, the sows can move freely around the pen 20, going at will to the feeder 21, the drinker 22, or to the resting area. The boar cage 23 is fixedly fixed in place in an open fenced area, where the sows can freely move from place to place and also approach the boar, which is placed separately in its own cage in this fenced area. In this case, the camera 24 and the microphone 25 are mounted on the cage in such a position where they face the sow when she approaches the cage with the boar in front of the boar. With this approach, a series of images of the sow can be obtained as she approaches for using these images in the analysis of the sow's movements, as shown in Fig. 3. As shown in Fig.2, short dividing walls 25A and 25B are installed, projecting forward from the cage with the boar on either side of the sow when she stands in front of the boar, so that signals from only one animal are received and analyzed separately from other animals located in the same general area (pen).

[0092] Microphone 25 is an elongated directional microphone designed to primarily pick up sounds from a sow immediately adjacent to the boar pen and reject sounds from other areas of the pen, such as other sows.

[0093] As shown in Fig. 3, the control system is capable of distinguishing between situations where the determination of the estrus period requires simple analysis and situations where complex analysis is required; in the simple analysis, an output signal indicating the determination of the estrus period is generated, and in the complex analysis, an output signal requiring additional intervention is generated before the determination of the estrus period can be made. In other words, the system can generate and indicate the level of confidence in the result, which allows the system to immediately proceed to the determination of the estrus period or indicate that the confidence is low and additional data is required. For example, additional data can be provided in response to a request based on the methods described below or based on other intervention by the worker. As another example, the worker can be used to press the dorsal region of the sow while collecting sounds and / or images.

[0094] The result of performing such a determination is that insemination is performed after receiving a signal indicating that the final determination has been made, so that the success rate of the insemination procedure is increased.

[0095] According to another method, as shown in Figs. 4-7, a pen 30 is created in which a group of sows are located, wherein these sows are potentially in a state of estrus. This pen 30 has two separate sections 31 and 32, where a sow can enter to eat from a feeder 33 or drink from a drinker 34. These sections are usually constructed so that only one sow can enter the section, and other sows cannot enter the rear part of the section. Each section carries a fixed camera 35, operating to periodically capture images of the rear profile of each sow when this sow enters the section. Since all sows must eat periodically, images of all these sows will be captured by the imaging system throughout the day. In this case, each sow approaches camera 35 at random times and its image can be captured each time it enters the area under consideration.In one embodiment, in such a situation, it is necessary to apply an identifier to the animal being imaged so that it can be identified for insemination. Therefore, in this case, animals can be tagged using a readable tag, such as an ear-worn RFID tag, detected by tag reader 37 in section 31.

[0096] As shown in the logical diagram shown in Fig. 7, where the steps of the method are shown in detail, the control system receives signals from the cameras and performs a series of steps, as shown in the diagram, to analyze the images to determine the estrus state.

[0097] Thus, each rear profile image shown in Fig. 5 is analyzed using the control system as shown in Fig. 7 to determine the region in the image as shown in Fig. 6 showing the vulva of a sow.

[0098] In some cases, the logic may differ slightly depending on the specific method used for image analysis. One method may not require selecting the vulvar region from a larger image before selecting suitable images. Selecting suitable images for analysis may be based on other criteria, such as the angles at which the images were taken, the orientation of the animal when the images were taken, and so on, to ensure that the best image is used for analysis.

[0099] Thus, images of the vulval region shown in Fig. 5 are analyzed to determine whether the sow is in a period of heat suitable for insemination based on features related to the shape, size, and / or color of the vulva. As noted above, this can be done after suitable images are selected to extract the desired region from a larger image. Alternatively, the entire image can be analyzed without selecting any region from it. Suitable images for analysis can be selected to discard some that are unsuitable due to angles, orientation, or position of the animal, depending on the method by which these images were obtained.

[0100] Then generate an output signal indicating that the estrus period has been determined, so that insemination is performed based on the reception of this signal.

[0101] According to a preferred embodiment of the method, the control system performs analysis using a program that compares the received data with previously collected data showing the vulval region of several sows known to be in heat. Thus, the program running in the control system is a neural network or other machine learning algorithm that has been pre-trained to detect the presence of heat by analyzing a large number of images of the vulval region of animals known to be in heat and animals known to be not in heat. This pre-trained program is stored in the control module for application to new images to be analyzed.Or it could even be a classical algorithm that is not a machine learning algorithm, using classical machine vision methods.

[0102] The first stage of the analysis preferably involves acquiring and analyzing multiple posterior profile images to select one or more suitable images that best depict the area of ​​interest. The control system uses a program to select suitable images from the entire set and to use the selected images to determine the estrus period. As part of this analysis, in some cases, the system must select only the relevant region of the vulva from the entire posterior profile image, as part of the algorithm requires first identifying this region and selecting it from the image component to enable further analysis. The system then evaluates each image to determine whether it indicates a positive or negative result.Once this determination is made, a comparison of all positive and negative results will yield a final determination of the condition and a confidence level indicating how many results agree. This information can be made available to assess whether insemination should be performed or whether further analysis with additional data is required.

[0103] One analysis that can be used for image qualification is based on the analysis system's determination of vulvar cleanliness. In this case, the farmer can be alerted if the vulvar area is too dirty and therefore has the potential to produce a poor result. In such cases, the images can be discarded, and the farmer can be alerted that action is required to clean the area before a suitable, accurate determination can be made.

[0104] In the next stage of analysis, size determination can be made not based on actual measurements, but by comparing the vulva size to a definable area of ​​the sow's posterior profile, such as the rump size or the size of the entire animal visible in the photographed rear view or posterior profile. This analysis can thus determine the ratio of the areas of the two regions captured by the imaging system and identify changes in this ratio that occur as the animal enters estrus.

[0105] After completing the first stage of analysis to select from the entire set of posterior profile images a subset of images that are the best or most suitable for demonstrating the vulvar area, these multiple images of the region of interest are selected and compared to determine the estrus period using statistical analysis.

[0106] In the alternative configuration shown in Fig. 1, camera 50 is positioned at the rear of section 11 to capture images of the animal in this section, both with and without the presence of a boar on the trolley. Rear profile images are preferably captured while the animal is standing, as this ensures that the vulval region is correctly represented in the rear direction and that this vulval region is in a specific position in the image, making it easier to select this region. This position thus ensures that parts of the region under consideration are represented in the image in a consistent manner, allowing for more efficient analysis.

[0107] The camera 50 and the associated control system thus operate to acquire or select an image when the animal is standing. In other words, the animal can be monitored in the section shown in Fig. 1 to determine when the animal is standing for capturing images, or the stimulating element 51 mounted on the wall of the section can be used to physically stimulate the animal to move to a standing position for capturing images. Alternatively, images can be captured periodically and simply discarded if the system determines that the animal is not standing.

[0108] When there is no identification information on an animal and, as shown in Fig. 4, this animal is free to move around the pen, then after the estrus period is determined, a mark for subsequent insemination can be applied to the sow for which the estrus period is determined in section 32 using a marker 36. Such marking systems by spraying paint or stamping are known and publicly available and can be used to mark the sow to indicate that it has been determined that she is in estrus.

[0109] It should be noted that the moment of insemination may be delayed relative to the analysis that determines the estrus period. The actual time of estrus may in some cases be delayed relative to the analyzed and detected visible changes in the vulva's appearance. In fact, the vulva's appearance usually changes some time before the onset of the active estrus reflex (which is commonly used to determine the estrus period in sows), and sometimes this change in appearance persists for several days after the cessation of the active estrus reflex. Therefore, the appearance of the vulva is a good indicator of the presence of estrus, but is not necessarily a good way to determine the actual moment for insemination. However, its appearance can be used to confirm that estrus is approaching or has begun.Another system, such as the image / sound analysis method described above, can then be used to better determine the appropriate time for insemination. Alternatively, a vulvar imaging system can be used to alert the sow to heat events that would be preferable to avoid insemination. Detecting such heat events is particularly important when working with young or new sows, as the farmer may want to inseminate them during their second heat, thereby skipping the first heat to ensure the animal's maturity. Therefore, it is important to detect the first and second heat events and not miss them, as otherwise the animal will become too old at the first insemination, reducing fertility throughout its life.

[0110] Because the invention as described above can be subject to various modifications and many apparently widely different embodiments of the invention can be made within the spirit and scope of the invention without departing from the spirit and scope of the invention, it is intended that all matter contained in the accompanying description be construed as illustrative only and not in a limiting sense.

Claims

1. A method for generating an output signal indicating whether a sow is in heat, comprising: defining a pen containing a group of sows, wherein the sows in said group are potentially in heat; periodically obtaining an image of the vulva of a sow from said group of sows; analyzing the obtained image of the vulva using a control system to conclude, based on the image, whether the sow is in heat;a database is formed from a set of images showing the vulva of a set of sows known to be in heat and images showing the vulva of a set of sows known to be not in heat, wherein said conclusion is based solely on a program using the obtained image in comparison with said database, and an output signal is generated indicating whether a sow is in the first heat in response to said conclusion.

2. The method of claim 1, wherein said analysis is performed by a control system using a program, wherein a large number of images showing the vulva of sows known to be in heat and sows known not to be in heat are first analyzed to generate data based on these previous images in order to create a program that, using this data, generates a signal to make said determination.

3. The method according to paragraph 1 or 2, characterized in that said conclusion is based on features that include the texture of the vulva or wrinkles on it.

4. The method according to any of the preceding claims, characterized in that said conclusion is based on features that include the height or width of the vulva.

5. A method according to any one of the preceding claims, wherein said determination is based on features that include a comparison of the size of the vulva with a defined area around the vulva of the sow.

6. The method according to any of the preceding claims, characterized in that said conclusion is based on features that include the color of the vulva.

7. The method according to any of the preceding claims, characterized in that said conclusion is based on features that include outwardly directed protrusions of the vulva.

8. The method of any one of the preceding claims, wherein an image analysis is performed to select for analysis an area in the image showing the vulva of the sow.

9. The method according to claim 1, characterized in that the images of the posterior profile are analyzed in order to select one or more suitable images that best show the vulva.

10. A method according to any of the preceding claims, characterized in that the sows are located in individual pens and the images are obtained using a camera mounted on the pens.

11. A method according to any of the preceding claims, characterized in that the sows are not isolated in individual pens, but the images are obtained by means of a camera installed in a place that is frequently visited by the sows.

12. A method according to any of the preceding claims, characterized in that a system for identifying the sows whose images are obtained is provided.

13. A method according to any of the preceding paragraphs, characterized in that after the period of estrus has been determined, a mark is placed on the sow in relation to which this determination has been made for subsequent insemination.

14. A method according to any of the preceding claims, characterized in that the control system is part of an autonomous system containing a camera, a processor for performing said determination, and an interface for communication with the worker.

15. The method according to any of the preceding claims, characterized in that the control system transmits data relating to said determination for a set of sows to the sow housing system.

16. A method according to any of the preceding paragraphs, characterized in that the images are taken manually, the worker holding the device for taking pictures.

17. The method of any one of the preceding claims, wherein a boar is provided which is isolated in a separate pen and pen, wherein said pen and pen are arranged such that sows in the pen are able to come into close contact with the boar when the boar or sow is moved.

18. The method of claim 17, wherein at least one collector device is provided for collecting images and / or sounds from each sow when it is in said close contact, and using said control system, the images and / or sounds are analyzed to determine whether the sow is in heat.

19. A method according to any of the preceding claims, characterized in that the control system is capable of distinguishing between situations where the determination of the period of estrus requires a simple analysis and a complex analysis, wherein in the simple analysis an output signal is generated indicating the determination of the period of estrus of the sow, and in the complex analysis an output signal is generated that requires additional intervention before the determination of the period of estrus is made.

20. The method according to any of the preceding claims, characterized in that after receiving said output signal indicating that the determination has been made, insemination is carried out.

Images